Qwen-3-30BA-3B

The 'Qwen3-30BA3B' dataset/benchmark is used to evaluate the performance of large language models (LLMs) in the context of inference on various hardware platforms, specifically focusing on the efficiency of matrix execution with CPU and scalable matrix extensions.

Papers using Qwen-3-30BA-3B (4)

SMEPilot: Characterizing and Optimizing LLM Inference with Scalable Matrix Extensions2026JudgeRLVR: Judge First, Generate Second for Efficient Reasoning2026Post-Trained MoE Can Skip Half Experts via Self-Distillation2026Pruning and Distilling Mixture-of-Experts into Dense Language Models2026